a3-highgpu-2g specs and pricing | GCP

GCP a3-highgpu-2g - Accelerator Optimized: 2 NVIDIA H100 GPU, 52 vCPUs, 468GB RAM. Available in 16 regions starting from $16029.42 per month. A -91% cheaper alternative is available.

Instance Type	a3-highgpu-2g
Details	Optimized for accelerated high performance computing workloads. a – Accelerator optimized 3 – Generation highgpu – High amound of GPU memory - 40x RAM/GPU ratio 2g – Number of GPUs
Instance Family	Accelerator optimized
Description	Accelerator Optimized: 2 NVIDIA H100 GPU, 52 vCPUs, 468GB RAM
vCPUs	52
Memory	468 GiB
CPU Architecture	x86_64
Processor	Intel Sapphire Rapids
Shared CPU	no
GPU	yes
GPUs	2
GPU Name	nvidia-h100-80gb
GPU Total RAM (GiB)	160
Sustained Discounts (SUDs)	no

Sole Tenancy	yes
Nested Virtualization	no
Custom Shapes	no
Confidential Compute	no
Maximum Persistent Disks	128
Maximum Persistent Disks Size	524288 GB
Disk Interface Type	NVMe
Local SSD	yes
Max Local SSD	6000 GB
Extreme Persistent Disks	no
Hyperdisk Extreme	yes
Hyperdisk Throughput	yes
Network Interfaces	gVNIC

Regional Availability

North America
California / us-west2
Iowa / us-central1
Mexico / northamerica-south1
Nevada / us-west4
Ohio / us-east5
Ontario / northamerica-northeast2
Oregon / us-west1
Québec / northamerica-northeast1
South Carolina / us-east1
Texas / us-south1
Utah / us-west3
Virginia / us-east4

Europe
Belgium / europe-west1
England / europe-west2
Finland / europe-north1
France / europe-west9
Germany / europe-west10
Germany / europe-west3
Italy / europe-west12
Italy / europe-west8
Netherlands / europe-west4
Poland / europe-central2
Spain / europe-southwest1
Sweden / europe-north2
Switzerland / europe-west6

APAC
Australia / australia-southeast1
Australia / australia-southeast2
Hong Kong / asia-east2
India / asia-south1
India / asia-south2
Indonesia / asia-southeast2
Japan / asia-northeast1
Japan / asia-northeast2
Singapore / asia-southeast1
South Korea / asia-northeast3
Taiwan / asia-east1

Middle East
Israel / me-west1
Qatar / me-central1
Saudi Arabia / me-central2

Africa
South Africa / africa-south1

South America
Brazil / southamerica-east1
Chile / southamerica-west1

Regional Prices

US Dollar ($)

Per Hour

On Demand

Geography	Region Name	Region	Savings Options	Instance Price
North America	Iowa	us-central1	compare	21.9581
North America	Oregon	us-west1	compare	21.9581
North America	Virginia	us-east4	compare	21.9581
Hidden	Hidden	hidden-1	compare	22.8364
Hidden	Hidden	hidden-2	compare	24.1781
Hidden	Hidden	hidden-3	compare	24.1781
Hidden	Hidden	hidden-4	compare	24.7314
Hidden	Hidden	hidden-5	compare	24.7314
Hidden	Hidden	hidden-6	compare	25.4253
Hidden	Hidden	hidden-7	compare	25.4714
Hidden	Hidden	hidden-8	compare	25.9106
Hidden	Hidden	hidden-9	compare	26.3739
Hidden	Hidden	hidden-10	compare	27.4476
Hidden	Hidden	hidden-11	compare	27.9543
Hidden	Hidden	hidden-12	compare	28.3888
Hidden	Hidden	hidden-13	compare	31.4471

Best AI models you can run on this instance

Top open-weight chat models ranked by Intelligence Index that fit in a3-highgpu-2g's 160 GB of GPU memory, assuming the model is sharded across all 2 GPUs. VRAM is estimated from the parameter count at the selected quantization, so treat it as guidance rather than a guarantee.

Chat Models

FP16 (full precision)

Model	Creator	Intelligence	Total params	Active params	~VRAM
Qwen3.6 27B	Alibaba	37.1#49	27B	—	65 GB
Qwen3.5 27B	Alibaba	33.8#69	27B	—	65 GB
Qwen3.6 35B A3B	Alibaba	33.0#81	35B	3B	84 GB
7 more models fit on a3-highgpu-2g Unlock the full ranked list and FP8 / INT4 quantization with a CloudPrice subscription.

Similar Alternative Instances

Find Similar Instances In:

Instance Type	Region	vCPUs	Memory (GiB)	Instance Price	Price Diff %
g2-standard-24	Iowa (us-central1)	24	96	2.0008	-91%
a2-highgpu-2g	Iowa (us-central1)	24	170	7.3468	-67%
a2-ultragpu-2g	Iowa (us-central1)	24	340	10.0554	-54%
a3-highgpu-1g	Iowa (us-central1)	26	234	10.9791	-50%